A voice-enabled virtual assistant may interact with a user using voice prompts to complete various activities. A user may interact with the voice-enabled virtual assistant by using specific commands or natural language phrases. The voice-enabled virtual assistant may respond by outputting audio such as text-to-speech output that may convert text based output to voice. The speech output of the voice-enabled virtual assistant may have characteristics that define its personality such as tone, timbre, sex, accent, etc. People may respond differently to different personality characteristics of a speaking voice.